Laser Spectroscopy and
Nanoparticle Research at The University of Texas in Austin |
||
Nanoparticle Generation by LAM Recently we received a patent for the Laser Ablation of Microparticles (LAM) process to make nanoparticles of a wide variety of materials (metals, semiconductors, and dielectrics). In the LAM process, a high-energy laser pulse hits a microparticle (typically 2-20 mue dia.), initiating breakdown and shock- wave formation. As the shock passes through the microparticle, it converts a high percentage of the mass to nanoparticles (<100 nm dia). Since the nucleation of nanoparticles follows the shock as a traveling wave, it is energetically efficient because the absorbed laser energy is only about 10% of the microparticle's heat of vaporization. We focus on generating small quantities of nanoparticles of a wide range of materials to demonstrate the viability and controllability of the LAM process and building flowing reactors capable of producing and collecting large quantities of nanoparticles. Nanoparticles have many potential applications because of their unique physical properties. Catalysts and low-temperature sintering materials take advantage of their larger surface to volume ratio, new optoelectronic materials result from the spatial confinement of electronic and vibrational excitations, and magnetic and ceramic materials are of higher quality because of their smaller grain sizes. However, the large-scale industrial application of nanoparticles requires further development of high-volume production methods. In experiments, we found that the LAM process is distinguished by nanoparticle distributions with a controllable mean diameter and a small dispersion (standard deviation / diameter) compared to nanoparticles generated by other processes, especially laser ablation from flat solid surfaces. In summary, the unique features of the LAM process are that the nanoparticles (1) are narrowly distributed in diameter, (2) have a mean diameter that can be controlled, (3) are as pure as the feedstock material, (4) preserve composition of the feedstock material, (5) are non-agglomerated, (6) can produce nanoparticles of virtually all solids, and (7) can be scaled to the production of large quantities. The as-manufactured distribution of sizes (Deltad/d ~ 0.20-.50) can be filtered using inertial impactors to produce a distribution of particles with Deltad/d = 0.2. For additional size selection, surfactant collected nanoparticles can be sorted to Delta d/d < 0.05 using size selective precipitation. A schematic of the reactor is shown in the left Figure. Microparticles are captured in a stream of gas at atmospheric pressure in the powder-aerosol generator that produces a sufficient particle number density (~108 cm-3) to absorb a significant fraction of the excimer laser energy (248 nm). To maintain laminar flow in the laser interaction cell and to provide a windowless design for the laser, the aerosol is focused by a flowing boundary gas after it leaves the nozzle. The laser light is brought to an elliptical focus at the end of the nozzle. Though the laser is pulsed, the laser repetition rate, the aerosol velocity, and the laser focal width down-stream are controlled so that microparticles just refill the focal volume in the time between laser shots. The nanoparticles are separated by a skimmer and sent though a filter (virtual impactor ) that separates any unablated or larger particles from the desired nanoparticle flow. This is particularly useful for materials that may produce bimodal size distributions. Our best results to date have yielded production rates for silver nanoparticles (dia. = 5 to 10 nm) of about 10 gm/hr. We are currently limited to a laser absorption of ~6% by the laser focus geometry, which was designed for experimental studies at laser fluences as large as 30 J/cm2. By changing the geometry of the laser focus, we can engineer greater absorption of the laser light and significantly improve the production rate. The figure shows typical examples of transmission electron micrographs of silver nanoparticles produced in helium at a fluence of 2.4 J/cm2 for varying pressures. a) 0.5 atm, b) 1 atm, c) 2 atm. This indicates that pressure is an excellent parameter for tuning distributions to a desired size. back to homepage |